Steroid signalling in the human ovarian surface epithelium wound healing
View/ Open
Date
2009Author
Papacleovoulou, Georgia
Metadata
Abstract
The human ovarian surface epithelium (hOSE) is a cell monolayer that covers
the surface of the ovary. Natural events like incessant ovulation, associated
reproductive hormone action prior to and post-ovulation, along with the ovulationassociated
inflammation, that result in injury and repair of hOSE, are considered to
have a role in the development of epithelial ovarian cancer (EOC). Progesterone is
apoptotic and anti-inflammatory, whereas androgens appear cytoproliferative for
hOSE. Local generation of these steroid hormones is subject to 3β-hydroxysteroid
dehydrogenase (3β-HSD) activity. Moreover, action of these hormones is achieved
through coupling to their cognate receptors, progesterone (PR) and androgen
receptors (AR). The overall aim of this thesis is to elucidate in vitro the regulation of
progesterone and androgen biosynthesis and downstream signalling during the injury
and repair of primary hOSE cells that were collected from pre-menopausal women
who underwent surgery for benign gynaecological disorders. Injury was mimicked
by treatment of cells with several pro-inflammatory cytokines, whereas repair was
mimicked with T-lymphocyte, ‘anti-inflammatory’ cytokines.
Immunohistochemical studies showed immunodetectable 3β-HSD in the
human ovarian cell surface of whole ovary and three-week cultured hOSE cells,
establishing 3β-HSD expression in vivo and in vitro. Cross-reaction of the 3β-HSD
antibody with both enzyme isoforms did not allow investigation of isoform
expression pattern. However, mRNA transcriptional studies with isoform specific
primers and probe sets for semi-quantitative (sq) and quantitative (q) PCR revealed
expression of both isoforms in hOSE cells; 3β-HSD1 mRNA was expressed at higher
levels relative to 3β-HSD2 mRNA in accordance with the preference of this isoform
in peripheral non-steroidogenic tissues.
Of the cytokines tested, only IL-1α and IL-4 affected 3β-HSD expression. IL-
1α suppressed 3β-HSD1 mRNA, whereas it up-regulated 3β-HSD2 mRNA as
assessed with qPCR, without though affecting total 3β-HSD protein and activity levels as assessed with western immunoblotting and radiometric activity assays,
respectively. IL-1α did not affect AR or PR mRNA levels, suggesting a balance in
androgen and progesterone biosynthesis during post-ovulatory wounding. IL-4
massively induced 3β-HSD1 and 3β-HSD2 mRNA and total 3β-HSD protein and
activity. It also attenuated AR mRNA and protein, without affecting PR mRNA.
Collectively, these data demonstrate that IL-4 sustains progesterone rather than
androgen signalling and this may be part of the anti-inflammatory steroid action that
protects hOSE from genetic damage. IL-1α effects appear to be mediated by NF-κB
signalling pathway. PI-3K and p38 MAPK appeared involved in IL-1α-induced 3β-
HSD2. IL-4-induced 3β-HSDs required STAT-6 and PI-3K pathways and also p38
MAPK at the case of 3β-HSD2. IL-4-attenuated AR was reversed by a p38 MAPK
inhibitor. These data suggest that steroid signalling by IL-1α and IL-4 involve
multiple signalling pathways.
In primary EOC, 3β-HSD1 and 3β-HSD2 transcripts were attenuated relative
to hOSE cells, suggestive of an acquired feature of neoplastic transformation.
However, both transcripts could be restored after IL-4 treatment, attesting a
therapeutic advantage of this cytokine.
In conclusion, we have shown that 3β-HSD is under inflammatory control
during ovarian post-ovulatory wound healing of hOSE. IL-1α- and IL-4-mediated
3β-HSD1 and 3β-HSD2 are regulated by multiple signalling pathways. Also, IL-4
was identified as an anti-inflammatory agent in hOSE with putative therapeutic
benefit in malignancy.